With the ongoing coronavirus outbreak, you may be finding yourself overwhelmed not only by the amount of information online, but also by all the different things that media are reporting about it.
What’s true, and what isn’t? Or you might not even know the basics of how coronavirus started, where it’s already spread to, and how it spreads.
Well, there’s now an easier way to educate yourself on it as well as stay up to date with the latest coronavirus news via CoronaTracker.
CoronaTracker collects data and aggregates news articles from trusted sources, and we spoke to the man who led its creation, Dr. Lau Cher Han.
Teamwork Makes The Dream Work
Dr. Lau is also the founder and CEO of LEAD, a startup that provides data science courses both online and offline.
As a chief data scientist himself, he realised he could do something to make data about coronavirus more accessible to the public.
He explained, “After we used the John Hopkins University (JHU) dashboard, we realised that it is very hard to gather data for in-depth analysis (for example, time series to monitor the growth, spread of the virus).”
“The detailed data we want is everywhere. It is also difficult to find localised news (related to Singapore, Malaysia, or Kuala Lumpur only). So, we built a platform and used technology to consolidate that information, and a portal to display the information and make the data accessible to the general public.”
He did this with the help of a team of over 500 people from all over the world including Malaysia, Australia, Singapore, US, Philippines, Taiwan, Japan, Netherlands and France, to name a few.
- Data scientists who follow the OSEMN framework to collect, clean, explore, data, and run their own analysis and present it using visualisations and dashboards, and who use Natural language processing to analyse content and identify meaningful topics,
- Full stack developers to help build the web and mobile applications,
- UI/UX designers to make sure the design is user-friendly, and that information architecture and flow is correct,
- Medical doctors and domain experts to help validate and verify information,
- The general public as testers for constant improvement of the platform.
“We wrote our scraper using Python to scrape data and news articles from websites like JHU and Tencent, and from news outlets like Reuters and BBC,” Dr. Lau added.
Ready In No Time At All
Work on the platform was only started on the morning of January 25 (also the first day of Chinese New Year), and the first version of it was ready in about 24 hours after that.
“We polished it, made sure data sources are connected and that news articles are filtered, to the best of our efforts. And then we launched it on January 27 via a Facebook live session,” Dr. Lau shared.
“We manually select the sources for news articles and data, and leave it to the machine to do its job. We will check it as soon as users report the numbers are not tallied.”
Such a tool is definitely useful, but we also had to ask Dr. Lau how the team ensured that the sources CoronaTracker uses are reputable.
To that, he replied, “Even trusted sources cannot be 100% sure that their news or their published numbers are accurate.”
“We have medical professionals and volunteers from government agencies who advise us on the best practice, and we always discard the information if we cannot get everyone’s agreement.”
So, rest assured that the team has the public’s best interests at heart, and Dr. Lau even shared that they will continue to improve it for a couple of days until they have incorporated all the initial features and handled all feedback from the public.
At the moment, what you’ll find on CoronaTracker are handpicked sources, analytics on coronavirus, prevention information, information on what coronavirus is, the latest news on it, and screening centres near you (at the moment, for Singapore and Malaysian states only).
If you’re reading this on another day, chances are that you might see even more features, as the platform is continuously being improved upon at an impressive speed.
As for what’s next for CoronaTracker, Dr. Lau said, “I am hoping to hand it over to some organisations to continue building it. After all, this is an open-source project, so all source code and publishable data are available on our Github.”
- You can read more on what we’ve written on coronavirus here.
Featured Image Credit: Dr. Lau